CRSS Coding Operations Resource Center
ATTENTION ALL OFFERORS: Crash Report Sampling System (CRSS) Coding Operations Resource Center #693JJ920R000070 As a result of an internal oversight with the processing of Questions, an Additional Ques... ATTENTION ALL OFFERORS: Crash Report Sampling System (CRSS) Coding Operations Resource Center #693JJ920R000070 As a result of an internal oversight with the processing of Questions, an Additional Questions document (Additional Questions_CRSS Coding_NHTSA Response_FINAL) has been posted in reference to the subject soliciation. Please reference NHTSA's response to these additional questions in the continued development of your proposal response. ********************************************************************************************************************************************** Amendment 0001 for Crash Report Sampling System (CRSS) Coding Operations Resource Center NHTSA issues the subject amendment to Solicitation #693JJ920R000070 (Crash Report Sampling System (CRSS) Coding Operations Resource Center) to update and correct the RFP as required.. Additionally, this amendment extends the proposal due date from Monday, November 9, 2020 at 1:00pm, EST to Friday, November 27, 2020 at 1:00pm EST. Please reference the attached Technical and Cost Question and Answers documents for NHTSA's response to Offeror questions, as well as the amended RFP for a detailed listing of RFP updates. ******************************************************************************************************************************************* The National Highway Traffic Safety Administration (NHTSA) is charged with the responsibility of reducing the personal and property losses resulting from motor vehicle crashes. Many sources of information are needed to permit researchers to adequately measure the characteristics of the highway safety environment. NHTSA data are essential to reducing the human and economic cost of motor vehicle crashes. NHTSA’s various data systems are the only source of real-world crash data for conducting basic research, identifying problem areas, developing effective countermeasures, identifying program and rulemaking needs, developing and evaluating programs, rules, and standards, evaluating new technologies, and providing information to accurately assess and allocate grants for reducing crashes. This statement of work applies to the record-based data collection, the Crash Report Sampling System (CRSS). CRSS is a sample of police reported crashes involving all types of motor vehicles, pedestrians, and cyclists, ranging from property-damage-only crashes to those that result in fatalities. CRSS is used to estimate the overall crash picture, identify highway safety problem areas, measure trends, drive consumer information initiatives, and form the basis for cost and benefit analyses of highway safety initiatives and regulations. CRSS obtains its data from a nationally representative probability sample selected from the estimated 5 to 7 million police-reported crashes that occur annually. By focusing attention on police-reported crashes, CRSS concentrates on those crashes of greatest concern to the highway safety community and the general public. The CRSS has three stratification stages: 1) Primary Sampling Unit (PSU), 2) Police Jurisdiction (PJ) and 3) Police Crash Report (PCR). At the first stage, the CRSS PSU sample, 3,117 counties in the country are grouped into 707 primary sampling units (PSU). A CRSS PSU is either a county or a group of counties. The 707 PSUs in the PSU frame are stratified into 50 strata by the four Census regions, urbanicity, vehicle miles traveled, total number of crashes, total truck miles traveled, and road miles. First, 101 PSUs are selected using a stratified probability proportional to size (PPS) sampling method. Then a sequence of sub-samples is selected from the original 101 PSU sample and strata are collapsed if necessary. This produces a sequence of nested PSU samples with different sample sizes selected from the collapsed strata. This sequence of nested PSU samples provides NHTSA flexibility to change and scale the PSU sample size in the future without reselecting the sample. Therefore, the final PSU sample is the result of a multiphase sampling mechanism in which the PSU selection probability is still approximately PPS. Sixty (60) PSUs were selected from 24 PSU strata as noted in Table 1. The second stage, the PJ Sample, comprises police jurisdictions (PJs) or groups of police jurisdictions. Within each selected PSU, PJs are stratified into three strata by their measure of size (MOS) which is a combination of crash counts in six categories of interest. A Pareto sampling method is used to select PJ samples from each PJ strata. This method produces overlapping samples when the sample is reselected. This method also reduces the potential of changes to the existing PJ sample when a new PJ sample must be selected because of PJ frame changes. The PJ inclusion probability under Pareto sampling is approximately PPS. The final stage, the PCR Sample, is the collection of police crash reports. The CRSS Samplers obtain police crash reports from each selected PJ. During each collection, all new police crash reports accumulated since the last collection are sequentially stratified into ten (10) police crash report strata (see Table 2). These 10 strata are formed based on the results of NHTSA’s internal and public data needs assessments. From each stratum, a systematic sampling method is used to select the police crash reported sample. The sampling intervals are determined in such a way that the final weights are approximately equal for all the police crash reports in the same stratum with the ultimate aim of reducing the sampling variance for the domain estimates. The target annual sample size is approximately 57,000 PCRs to be selected for further coding. The CRSS data collection effort begins with sampling activities. The CRSS Samplers review the crash reports to review, stratify and list important sampling variables, such as PSU, Police Jurisdiction (PJ), Crash Date, Crash Time, Stratum and Special Crash Investigation (SCI) categories in the Police Accident Report Sampling Engine (PARSE). The PARSE application runs an algorithm to select cases to be coded for the annual CRSS data file. The CRSS Sampler may then upload the case materials into the PARSE application for further coding. The PARSE application routinely transmits the case materials for the selected cases to be coded. CRSS coding begins only after a PCR has been selected for the CRSS sample. There is a separate NHTSA contract, the Crash Report Sampling System (CRSS) Sampling Resource Center, that awards the responsibility to review, stratify and select cases. CRSS Samplers are trained to review and categorize the crash reports into three categories: NTS, CRSS applicable stratums or out of scope. The selected cases for NTS and CRSS are identified by the Police Accident Report Sampling Engine (PARSE) application and made available for coding by the CRSS Coders. CRSS Coders are trained data entry personnel who interpret the crash reports into the CRSS Records Based Information System (RBIS) application. The CRSS RBIS application allows for the CRSS Coder to review, code and perform quality control activities. Approximately 120 data elements are coded and added into RBIS. After coding, quality control is performed on the data to ensure validity and consistency with the FARS/CRSS coding and validation manual guidance. After coding is completed for all the selected cases, CRSS data files and coding documentation publicly available. This Contract supports the CRSS Coding Operations that include but are not limited to coding, quality control and technical guidance for the non-state run Crash Report Sampling System (CRSS) sites, which is 57 of the 60 sites.
Data sourced from SAM.gov.
View Official Posting »